A strong close to 2025 and an even faster start to 2026
2025 was a defining year for Picsart’s Creative APIs.
Over the course of the year, we shipped 11 monthly release cycles and delivered 30+ major improvements across GenAI, Image, Video, and Variable Data Content APIs. These releases included new endpoints, upgraded AI models, expanded format support (including HEIC/HEIF), and deeper video and automation capabilities.
Our focus was clear: help teams build scalable, production-ready creative workflows with less complexity and fewer integrations.
In 2026, we’re doubling down. We aim to move faster, improve quality even further, and continue building a complete creative platform that grows with your needs.
January 2026 updates
AI Providers Hub: one integration, 20+ image generation models
The biggest launch this month is the AI Providers Hub.
With the AI Providers Hub, you can now test and run over 20 image generation models through a single API, with almost no code changes. You can explicitly select a model or let Picsart automatically apply a strong default.
As new models become available across vendors, we’ll add them to the hub, removing the need for repeated integrations and migrations.
What this enables:
- Faster experimentation across multiple models
- Easier model switching without refactoring
- A future-proof GenAI setup that evolves with the market
👉 Learn more about the AI Providers Hub
Bonus (limited-time benefit):
All models in the AI Providers Hub currently run at a fixed price. Even premium models (often expensive when used directly) can be used for just 2 credits per generated image, making experimentation predictable and cost-efficient.
Smart Crop: prompt-guided image cropping
Traditional cropping requires pixel coordinates and often leads to important content being cut out. Smart Crop changes that.
With Smart Crop, you can use natural language prompts to define what should be included in the crop. The API understands intent and focuses on the relevant parts of the image.
This is especially useful when:
- Creating multiple product images from a single source photo
- Avoiding center-crop mistakes
- Reusing the same image for different SKUs or layouts
Selective Blur: blur faces and car plates in one API call
Blurring sensitive content used to require multiple steps and custom logic. Now it’s one API call.
With Selective Blur, you simply specify what to blur using a segment name, and the API handles detection and masking automatically.
Some examples of segments:
facescar plateforegroundbackgroundhat
This makes it easy to meet privacy and compliance requirements without building complex pipelines.
Remove Background v10: now the default model
We’ve upgraded the default Remove Background experience.
The latest model version:urn:air:picsart:model:picsart:sod@10
is now the default across the platform.
What to know:
- The previous version (
@9) is deprecated and removed - If you explicitly set v9, please update your integration
- If you don’t specify a model, v10 is applied automatically
The new version delivers better edge detection, cleaner cutouts, and more reliable results across complex scenes.
👉 Try the Remove Background API
Looking for the full change log?
👉 Browse the release notes
We’re just getting started.
Stay tuned for more updates and see you in February with the next wave of improvements.
